Influence of Ladle Addition Practice on the Castability of ...

International Journal of Mineral Processing and Extractive Metallurgy 2018; 3(1): 1-14

http://www.sciencepublishinggroup.com/j/ijmpem

doi: 10.11648/j.ijmpem.20180301.11

ISSN: 2575-1840 (Print); ISSN: 2575-1859 (Online)

Influence of Ladle Addition Practice on the Castability of Typical Low Carbon Low Silicon Steel

Anil Kumar1, *

, Shiv Kumar Choudhary1, Saroj Kumar Singh

2

1Research and Development, Tata Steel Limited, Jamshedpur, India 2Steelmaking and Casting Division, Tata Steel Limited, Jamshedpur, India

Email address:

*Corresponding author

To cite this article: Anil Kumar, Shiv Kumar Choudhary, Saroj Kumar Singh. Influence of Ladle Addition Practice on the Castability of Typical Low Carbon

Low Silicon Steel. International Journal of Mineral Processing and Extractive Metallurgy. Vol. 3, No. 1, 2018, pp. 1-14.

doi: 10.11648/j.ijmpem.20180301.11

Received: April 25, 2018; Accepted: May 14, 2018; Published: June 12, 2018

Abstract: Continuous casting of low carbon low silicon steel processed though Basic Oxygen Furnace (BOF) followed by

Online Purging station (OLP) and Ladle Furnace (LF) inherently suffers from the problem of clogging of submerged entry

nozzles (SEN) during casting due to gradual alumina deposition at the inner surface of the refractory wall. This restricts the

incoming flow of liquid steel from tundish to mould through SEN and thus limits the productivity of the caster. On the other

hand, excess alumina deposition and their periodic dislodgement from the nozzle refractory surface gives rise to undue stopper

rod movement and melt level fluctuations in the mould. All these abnormalities lead to decreased production of good quality

slabs. In the present work, influence of ladle addition practice on the castability of typical low carbon low silicon steel has been

investigated for a conventional thick slab caster. For identification of inclusions characteristics at different stages of

processing, liquid steel samples were collected using the lollipop and a specially designed large sampler during casting. Also,

the clog deposits in the used SEN and the natures of inclusions in the clog deposits as well as in the liquid steel were identified

by SEM–EDS analysis. Subsequently, to measure the process the different plant trials were carried out for several casting

sequences using different deoxidation practices. The upstream processing parameter and corresponding casting data with

nozzle clogging index (NCI) for a complete casting sequence were examined for evaluating their casting performances.

Addition practice of deoxidant and lime during the ladle treatment has been found to have an important effect of the liquid

steel casting. It has been clearly demonstrated the faulty addition practice can lead to severe nozzle clogging during casting.

Though a series of plant trials the best addition practice has been determined and implemented in the regular production in the

plant for ensuring the smooth production of the given grade of steel. For the quantification of steel cleanliness, total oxygen

content (T [O]) of liquid steel, was measured from the samples collected after each ladle treatment stage. T [O] of liquid steel

was found in the range of 60−120 ppm at OLP and 30−55 ppm at tundish of low carbon low silicon grade of steel.

Keywords: Al Killed Steel, Ladle Treatment, Slab Casting, SEM-EDS, Alumina, Inclusions, Nozzle Clogging,

Clogging Index

1. Introduction

Aluminium killed steel inherently suffers from the

problem of nozzle clogging during continuous casting [1, 2].

There is gradual deposition of residual fine alumina and other

solid non-metallic inclusions (NMI) at the wall of refractory

submerged entry nozzles (SEN), restricting the passage flow

of liquid steel through them during casting operation.

Therefore, nozzle clogging has been considered as one of the

important process upsets in the smooth production of Al-

killed steel. Alumina in liquid steel originates from residual

unfloated deoxidation products of ladle treatment and those

generated subsequently from reoxidation reactions of liquid

steel during subsequent transfer operations [3, 4]. Besides

these physical entrapments of solid exogenous particles and

those generated from unintentional side reactions between

liquid steel and refractory also give rise to additional solid

2 Anil Kumar et al.: Influence of Ladle Addition Practice on the Castability of Typical Low Carbon Low Silicon Steel

inclusions [5, 6]. Such steels are commonly produce through

three process routes viz., (a) direct route where liquid steel is

sent directly for the casting just after primary deoxidation

and argon rinsing (ARS) treatment in the ladle during tapping

from the vessel, (b) in the LF route where liquid steel is

further treated in the ladle furnace after ARS before casting,

and (c) RH route where liquid steel is first homogenised and

partially deoxidised after the vessel tapping and ARS then

subjected to treatment in the RH degasser before casting.

Amongst these routes, casting of LF treated steel has been

found to be more problematic in comparison to other routes.

During liquid steel casting, the severity of nozzle clogging is

commonly judged by monitoring the movement of tundish

stopper rod, mould level fluctuations (MLF) and frequency

of argon flushing. Undue MLF arising from the sudden

dislodgement of large alumina chunks from the nozzle

deposit hampers smooth casting operations, and leads to

downgrading of costly cast slabs besides limiting the caster

productivity.

Direct quantification of nozzle clogging is difficult as

nothing is visible during casting. Indirectly, nozzle clogging

index (NCI) has been widely adopted for judging the level of

liquid steel cleanliness and its casting performance [7]. For a

more realistic assessment a dynamic clogging index has been

developed to keep track of the nozzle clogging during casting

[8, 9]. Dynamic clogging index is based on the actual casting

variables such as: stopper rod position, tundish weight,

casting speed, etc. In this model, NCI data is continuously

recorded at every 5 second during casting and thus about

700−900 data are generated per heat depending upon the

casting duration. The slope of the NCI versus time data is

continuously calculated during casting. Subsequently, slopes

of the NCI plots statistically correlated with the upstream

steelmaking parameters for establishing their role in alumina

deposition in SEN during casting. In view of this present

work has been undertaken for establishing the prime causes

of clogging in low carbon Al killed grades and find out the

appropriate countermeasures for its minimisation. Work

involved collection of liquid steel samples from the online-

argon purging (OLP) station, LF and tundish using a

specially designed large size sampler and the routine lollipop

samplers. Further liquid steel samples were subjected to

inclusion characterisation using scanning electron

microscope and energy dispersive spectroscopy (SEM-EDS)

and total oxygen measurements. Deposited clog materials

from several SENs were also collected and examined for

identifying various clogging agents and their possible origin

in the upstream steelmaking processes. Finally, plant trials

were conducted with modified liquid steel deoxidation

practices and improvement in casability was evaluated on the

basis of the observed inclusion characteristics and

corresponding NCI values of various heats in the casting

sequence.

2. Experimental

Figure 1 schematically describes the outline of the

standard operating procedure (SOP) adopted for the

production of low carbon Al killed steel. Depending on the

specification of steel grades (e.g. a typical specification of

the grade is listed in Table 1) combination of aluminium and

different types of ferroalloys are added to the liquid steel

along with lime in proper sequence during their ladle

treatment. In the present work, for identification of inclusions

characteristics at different stages of processing, liquid steel

samples were collected using the routine lollipop and a

specially designed large sampler before and after OLP and

LF treatments, and from the tundish during casting. Design

of the special sampler is shown in Figure 2. After taking the

liquid steel sample using large sampler, it was sliced into two

halves by cutting centrally along the longitudinal as well as

transverse directions as shown in Figure 3. The shrinkage

cavity containing upper portion of the sample was discarded

and only its lower sound portion (Figure 3b) was used for

inclusion characterisation. Characteristics of inclusions in

liquid steel samples were examined using SEM-EDS [10].

Figure 1. Schematic SOP adopted to produce of low carbon low silicon steel.

International Journal of Mineral Processing and Extractive Metallurgy 2018; 3(1): 1-14 3

Table 1. Chemical specification of a typical low carbon low silicon grade of steel.

% C % Mn % Si % Al % S % P % Nb [N], ppm

0.07-0.1 1.0-1.5 ≅ 0.05 0.021-0.04 0-0.01 0.01-0.015 0.004-0.01 < 90

Figure 2. Design of large liquid steel sampler used in the present work.

Figure 3. Liquid steel sampling using large sampler (a) central longitudinal slice (b) transverse section.


After completion of the casting sequence the used SENs

were collected to examine the nature of inclusions in the

deposit materials using SEM-EDS. Collected SENs after

casting were broken and the deposited clog material were

taken from its bottom and port areas. Samples were

considered from various locations of the clog deposits and

the natures of inclusions were identified. Subsequently, to

measure the process the different plant trials were carried out

for several casting sequences using different deoxidation

practices. During the trials, the details of upstream liquid

steel processing and corresponding casting data were

collected. Also, NCI for a complete casting sequence was

examined for evaluating their casting performances.

Methodology of NCI calculation has already been reported

elsewhere [9].

For the quantification of steel cleanliness, total oxygen

content of liquid steel T [O] was measured for the samples

collected after each ladle treatment stage. Samples for this

measurement were taken from the routine lollipop samples of

liquid steel. The central thick portion of the lollipop was cut

into 3 pieces and cylindrical samples of 5 mm diameter and 8

mm length were made. Subsequently the cylindrical samples

were subjected to fine grinding and polishing. Weight of

cylindrical samples ranged between 0.5−1.0g. For avoiding

any contact with air prior to test those samples were

immediately dipped into acetone and stored in amber colour

glass bottles. The combustion analysis of the sample was

done within two hours of the polishing in order to prevent

oxygen pickup at the surface of the sample. During the

combustion analysis, if the first two values measured for the

total oxygen content were within 4-5 ppm of each other, their

average was taken as the final T [O] value for that sample,

otherwise the third piece of that sample was analysed. If all

the three values were within 10 ppm of the mean value, the

average of three was taken as final value for that sample.

However, if there was one reading exceedingly higher than

the other two, this reading was discarded and the average of

the other two reading was taken as final T [O] value. The

exceedingly high reading, whenever it occurred can be

explained by taking into consideration the possibility of

internal porosity in the sample which were not apparent by

the visual inspection of the sample.

3. Results and Discussion

3.1. Examination of Deposited SEN Clog Materials

Visual examination of used several alumina graphite SEN

after casting revealed that its barrel was almost clean and

practically free from any alumina build up. However, most of

the alumina deposition took place around the bottom of the

SEN (Figure 4). Bulk of the alumina deposit consisted of

friable chalky alumina particles. Samples of alumina build up

were collected from the various locations of the deposits and

examined using SEM-EDS. In general, coral shaped clusters

of pure alumina and platy spinel MgO.Al2O3 particles were

observed along with small quantity of other inclusions in the

nozzle deposits. Inclusion particles were heterogeneously

distributed throughout the deposits. Bulk of the deposits

consisted of round and angular pure alumina particles as

shown in Figure 5. Commonly, such angular alumina

particles form due to late addition of aluminium to the

already deoxidised liquid steel [11, 12]. Besides these,

globules of iron particles were also observed which were

entrapped in the alumina clusters at few locations (Figure 6).

Entrapped iron particle took the globular shape essentially

due to the non-wetting characteristics of alumina towards

liquid steel. Non-wetting alumina essentially prevented

entrapped liquid steel from spreading in the inter-particle

voids of the clusters. That is why entrapped liquid steel

particles got locked and froze within the clusters (as round

particles) due to insufficient capillary force available to

drive/ooze them out [13-15]. As already mentioned, bulk of

the inclusion deposits consisted mostly of coral shaped

clusters of pure alumina (Al2O3) and spinel (MgO.Al2O3)

inclusions, essentially indicating these two as the prime

clogging agents in the current ladle deoxidation practice [16-

19]. Common physical characteristics of above clogging

agents are presented in Table 2.

(a)

(b)

Figure 4. Sample collections (a) used SENs after completion of casting

sequence and (b) deposited materials collected from the bottom of used SEN.


(a)


(b)

Figure 5. Pure alumina clusters (a) and the spinel (b) found in the SEN clog deposits.

(a)


Figure 6. Entrapped globular iron particles found in the SEN clog deposits.

Table 2. Characteristics of common clogging agents in liquid Al-killed steel.

Parameters Alumina, Al2O3 Spinel, MgO.xAl2O3

State at 1600°C Solid Solid

Surface tension (w.r.t.

liquid iron at 1600°C)

Non-wetting

θc ~140°

Non-wetting

θc ~ 120°

Clustering tendency Strong tendency Weak tendency

Floatation in liquid steel Easier

(for large clusters) Difficult

Entrapment of liquid steel Liquid steel repellant; bulk of alumina

deposit practically free from frozen metal

Liquid steel repellant; clustures of spinal also practically free from frozen

metal

Presence Present in all liquid steel samples from LF,

Tundish & SEN clog Present in all LF, tundish & clog samples

Source Unfloated products of deoxidation &

reoxidation

Exclusively forms if Mg is present in liquid steel (as low as 0.5 ppm Mg is

sufficient to form spinel)

Reaction between dissolved [Al] in liquid steel & MgO present in ladle lining

Prolonged treatment & holding time (in MgO lined ladle)

Mg present in ladle additions

Overkilling of MgO containing ladle slag (very low slag FeO)

3.2. Inclusion Characteristics of Liquid Steel

Table 3 shows the distribution of total inclusion area

percentage in liquid steel samples at different stages of liquid

steel from ladle treatment up to casting. In order to ensure

that inclusions measured in the laboratory must represent

fairly well its content in total volume liquid present in the

ladle, all samples were collected using the large size special

samples. It was observed that most of oxide and sulphide

NMI were present in almost all the samples taken before and

after LF treatment in addition to some slag entrapment

observed in the tundish. Alumina and spinel inclusions were

mostly observed in all liquid steel samples. Isolated angular

alumina particles (size < 5µm) were more frequently seen.

Few compact alumina clusters were also present in some of

those samples.


Table 3. Variation of amount of inclusions from ladle furnace (LF) to tundish.

Type of inclusions observed Total area fraction of inclusion, percent

before LF treatment after LF treatment Tundish

Alumina-titania 0.00188 0.00138 0.00053

Alumina 0.00164 0.00191 0.00123

MnS 0.00285 0.00406 0.00695

MnS–associated with alumina 0.00698 0.00567 0.00699

MgO-Al2O3 0.00024 0.00475 0.00145

Complex Ca-Al-Si-Mn-O inclusion (slag) - - 0.002

Figure 7 shows a typical alumina clusters. In addition,

some exclusively reoxidation product Al2O3.TiOx was found

in liquid steel after OLP treatment. Al2O3 inclusions formed

first, causing local depletion of Al and supersaturation of Ti

followed by reaction Al2O3 (s) + [Ti] + x [O] = Al2O3.TiOx

(s) and Al2O3.TiOx got precipitated over preexisting solid

inclusions like alumina and spinel [20 – 24]. It has been

reported [25]

that the formation of MnS type of inclusion in

common low carbon Al killed steel would be possible only

during the solidification of liquid steel in the mould during

solidification where the concentration of [S] in the mushy

zone gradually increases and reach its saturation level.

Presence of MnS inclusion in the present study could be due

to relatively slow liquid solidification in the thick samplers.

Figure 7. SEM-EDS analysis of alumina clusters found in liquid steel samples.

Al2O3.MgO (Spinel) was considered to originate from the

side reactions especially from ladle refractory lining, top slag

and/or traces of Mg content of aluminum used for the liquid

steel deoxidation. In the present work, samples taken after

the ladle treatment contained exogenous CaO-SiO2-Al2O3

inclusions containing small quantity MgO and MnO which

were possibly originated from the slag entrapment. These

inclusions were liquid initially and got solidified on cooling

during the sampling process. In addition, some MnO–SiO2

containing only a small quantity of Al2O3 inclusions were

observed in few cases. Few isolated MnS particles or in

association with other oxides were also found, which formed

exclusively during the last stage of solidification, were also

present in the liquid steel samples. Entrapped slag particles of

different sizes were found almost randomly distributed in

various samples (Figure 8). Spinel normally forms when

liquid steel contains dissolved magnesium in trace quantity.

Thermodynamically, as low as 0.5 ppm [Mg] in liquid

aluminium killed steel is sufficient enough to produce spinel

inclusion [26, 27]. Spinel inclusion has a relatively weak

floatability in the liquid steel. Therefore, the best solution is

to avoid spinel formation by eliminating all possible sources

of metallic magnesium in liquid steel.


Figure 8. X-ray map of constituent elements of a typical slag entrapment.


3.3. Inclusion Size Distribution of Liquid Steel

Size distributions of inclusions were measured in order to

find out their relative populations in the liquid steel samples.

Size distribution of NMI in liquid steel at different stages

from the ladle treatment to caster is shown in Figure 9. It can

be seen that the smaller inclusions less than 5 µm were

relatively abundant in all samples. Most of the inclusions in

this range were found to be primarily the residual

deoxidation product. Reoxidation inclusions of sizes ranging

from about 2-10 µm were also observed. In comparison,

there were only a few large size inclusions. Even then their

relative population was small but such inclusions are

considered to be more detrimental to steel properties. They

occurred as clusters, aggregates, and sometimes as dendrites.

However, in some cases slag entrapment as large as 300

microns has also been observed in cast slab samples [28].

Figure 9. Inclusion size distribution of a typical heat at different stages at (a) LF before treatment, (b) LF after treatment and (c) tundish.


3.4. Plant Trials and NCI Distribution

As already indicated above, nozzle clogging during casting

was mostly due to presence of large number of unfloated fine

alumina and spinel inclusions in the liquid steel, therefore

attempts have been made to minimise their population by

taking measures in the upstream liquid steel processing and

handling operations. In the current practice, low carbon Al

killed steels are produced by using step wise deoxidation

process. Initially preliminary deoxidation is being carried out

by adding commercial aluminium bars to the crude steel

during tapping from the vessel. Further, Al wire is injected at

OLP station after that liquid steel is sent to the LF for further

treatment. In the LF depending upon the aluminium content of

further Al addition to liquid steel is made. In the present work,

plant trials were conducted for with modified deoxidation

practice in which most of the aluminium addition were made

during the vessel tapping and only a small quantity of the same

was added in the subsequent ladle treatment. Besides this, total

lime addition was also reduced as compared to previous

practice. Casting performance of all trial heats was evaluated

on the basis of observed NCI values. Actions taken in the

modified practices (trials) are shown in Table 4.

Table 4. Additions during ladle treatment of liquid steel.

Ladle additions Old practice Modified practice

trial 1 trial 2 measures taken for optimisation

Lime during tapping, kg 500-800 500-600 500-600 limit of maximum addition reduced by 200 kg; range narrowed down

Lime in LF, kg 300-500 400-500 400-500 lower limit increased by 100 kg; addition range restricted

Al bar at tapping, kg 200-400 300-320 320 reduced

Al-wire before LF treatment 100-150 0-60 0 as needed

The overall NCI distribution in different heats of two trial

casting sequences has been compared with the base data

(existing plant operation) in Figure 10. In general, most of

the trial heats exhibited relatively lower NCI values as

compared to the previous practice (Figure 10). Higher

aluminium addition during vessel tapping and only a small

addition in subsequent LF treatment gave relatively less

clogging problem during casting. This has been attributed to

the generation of mostly easily floatable large alumina

particles due to relatively higher oxygen content of liquid

steel during tapping and reduced generation of problematic

finer alumina inclusions subsequently. Avoiding the late

aluminium addition prevented generation problematic fine

alumina inclusions. During trials, final slag chemistry in the

LF remained almost same. With these modified deoxidation

process, on an average about 60 percent drop in NCI was

found in most of the trial heats (NCI ≥ 0.66 for bad heats in

previous case and NCI≅ 0.26 for good heats) during casting.

Figure 10. Comparison of NCI performance on deoxidation practices against each plant trial over a complete casting sequence

Figure 11 presents a typical online distribution of NCI

(blue line) in different heats for a trial casting sequence,

where less lime addition was made at the liquid steel tapping

and more of it was added in the subsequent LF treatment. It

can be seen that the number of argon flushing through

stopper rod (auto and manual) was significantly higher than

that of a general casting operation. Slag characteristics plays

important role in absorbing the floated alumina from the

liquid steel. In the plant trials slag formation was modified by

minimizing the quantity of lime addition during tapping.


Figure 11. On-line data of stopper rod movement, nozzle clogging index and tundish weight with progress of casting in a trial casting sequence of 12 heats.

Figure 12 shows the major constituents of the steelmaking

slag at different stages of ladle treatment. It can be observed

that CaO content in the slag after OLP treatment was about

45% in comparison to about 55% after LF treatment. The

excess or insufficient lime may also be detrimental for

clogging [29]. Lime addition during liquid steel tapping

(OLP) and LF treatment is shown in Figure 13. It can be

observed from both the figures (a) and (b) that there were a

wide range of variations (about 400–1400 kg per heat) of

total lime additions during Tap and LF treatment. In addition,

the contributions of lime additions at Tap and LF for such

grades were not consistent. The ratios of tap to LF additions

per heat were varied from 0.5 to 2.0 (can be seen in ‘b’ part

of the figures). Such types of variations of lime additions

may cause the performance of alumina holding capacity of

the slag which may affects the inclusion floatation and hence

affects NCI performance during the casting operations.

Figure 12. Main constituents of ladle slag after OLP and LF treatments of liquid steel.


Figure 13. Pattern of lime addition during liquid steel treatments: (a) total lime at Tap and LF and (b) ratio of lime Tap to LF.

3.5. Steel Cleanliness

Figure 14 shows the distribution of T [O] of liquid steel at

OLP and Tundish. In Al-killed steel with T [O] content of 30

ppm, the number of 2 µm size inclusions have been estimated

[30, 31] to be around 4.5 x 1012

, which is considerably large

enough to clog the nozzle. Therefore, it appears that even in

steels with very high cleanliness level (e.g., T [O] ∼ 10 ppm),

there still have enough inclusions to clog the nozzle. As a

matter of fact, bulk of the nozzle deposit consists of only fine

(2-5 µm) alumina particles, which again confirm the role of

microinclusions on nozzle clogging. As discussed already,

overall steel cleanliness was measured through its total

oxygen content using LECO (TC600) oxygen determinator. T

[O] content of steel is summation of dissolved oxygen and

the oxygen combined with oxide non-metallic inclusions. In

solid steel, dissolved oxygen is practically negligible.

Therefore, the measured T [O] essential is taken as an

indirect measure of the oxide NMI contents of steel.

Dissolved oxygen in liquid steel is commonly measured

using solid electrolyte sensors (CELOX). In liquid Al-killed

steel dissolved oxygen vary between 3-5 ppm (almost

constant).

Figure 14. Distribution of T[O] content of liquid steel at OLP and Tundish.

Due to the small population of large inclusions in the steel

and the small sample size for T [O] measurement (normally

1.0 g), it is rare to find a large inclusion in the sample. Even

if a sample has a large inclusion, it is likely discounted due to

its anomalously high reading. Thus, T [O] content really

represents the level of small oxide NMI only. A low T [O]

content, however, decreases the probability of large oxide

inclusions. Clean steel requires control of the size

distribution, morphology and composition of oxide NMI in

addition to the amount. It can be seen from the figure that T

[O] of liquid steel was found in the range of 60−120 ppm at

OLP and 30−55 ppm at tundish of low carbon low silicon

grade of steel.

4. Conclusion

Slab casting of low carbon and low silicon aluminium

killed steel has been quite problematic in comparison to other

grades. These grades suffered from undue stopper rod

movement and higher mould level fluctuation during casting,

leading to significant downgrading of cast slabs. Present

work aimed at identifying the potential causes of above and

recommends measures to the plant for the improvement.

Towards this, samples of clog material from several used

SEN were collected along with corresponding liquid steel

samples from the tundish and from the ladle after each stage

of treatments. Characteristics of the inclusions present in the

SEN deposits and liquid steel at LF and tundish were

determined using SEM-EDX in order to establish their roles

on the observed castability of liquid steel. Most of the

inclusion deposition was found at the bottom of SEN (port

region). Inclusion deposit consisted of white friable (chalky)

material. SEM-EDX examination revealed the clog material

contained mostly coral shaped alumina clusters and spinel

inclusions along with few globular frozen steel particles

inside the deposits. Corresponding liquid steel samples from

LF and tundish found to contained large number of finer (less

than 5micron size) solid alumina and spinal inclusions. In

addition, some reoxidation products and slag entrapments

were also observed. Wide variation in the quantity of Al

additions (200-400 kg Al bar during Tap and 100-200 m of Al

wire before LF treatment) and lime additions (about 400–

1400 kg per heat of total lime additions during Tap and LF

treatment) was observed, particularly during the primary and

secondary deoxidation of liquid steel in the ladle. Large

addition of aluminium and lime led to poor castability of the

steel. All these influencing parameters of ladle additions

practice were optimized through several plant trials for better


castability of liquid steel in terms to lower NCI values during

casting operations. Al additions during Tap was reduced to

300-320 kg and Al wire feeding was reduced to 0-60 m at

OLP. Similarly, the total lime addition was reduced to 900-

1000 kg. T [O] of liquid steel was found to vary over a wide

range (60−120 ppm) just after primary deoxidation during

tapping and OLP and after subsequent treatment it was found

to decrease up to 30−55 ppm in the tundish.

Acknowledgements

The author would like to thank Tata Steel to allow this

work to publish. Dr. T. K. Roy and Mr. V. Balakrishnan are

also acknowledged for helping during sampling through

special samplers.

References

[1] L. Zhang and B. G. Thomas: ISIJ Int., 43(2003), 271.

[2] L. Zhang, X. Wang, K. Cai and B. G. Thomas: Steelmak. Conf. Proc., ISS-AIME, 85(2002), 431.

[3] Y. Hara, A. Idogawa, T. Sakurya, S. Hiwasa and H. Nishikawa: Steelmak. Conf. Proc., Washington, 75(1992), 513.

[4] A. W. Cramb and M. Byrne: Steelmak. Conf. Proc., Warrendale, 69(1986), 719.

[5] V. C. Galindo, R. D. Morales, J. A. Romero, J. F. Chavez and M. V. Toledo: Steel Res., 71(2000), 107.

[6] O. Wijk: 7th International Conf. on Refining Process, MEFOS, Lulea, Sweden, (1995), 35.

[7] D. Bolger: Steelmak. Conf. Proc., Chicago, 77(1994), 531.

[8] N. U. Girase, S. Basu and S. K. Choudhary: Ironmak. Steelmak., 34(2007), 506.

[9] A. Kumar, G. M. Kumar, S. K. Ajmani and S. K. Singh: Ironmak. Steelmak., 44(2017), 210.

[10] R. Rastogi and A. W. Cramb: Steelmak. Conf. Proc., ISS, 84(2001), 789.

[11] W. K. Tiekink, A. Pieters and J. Hekkema: Steelmak. Conf. Proc., Chicago, 77(1994), 423.

[12] Y. Vermeulen, B. Coletti, B. Blanpain, P. Wollants and F. Haers: Steelmak. Conf. Proc., ISS, 83(2000), 175.

[13] S. Ogibayashi, M. Uchimura, Y. Maruki, D. Mizukoshi and K. Tanizawa: Steelmak. Conf. Proc., Toronto, ISS, 75(1992), 337.

[14] Y. Fukuda, Y. Ueshima and S. Mizoguchi: ISIJ Int., 32 (1992), No. 1, 164.

[15] J. Poirier, B. Thillou, M. A. Guiban, G. Provost: Steelmak Conf. Proc., 78(1995), 451.

[16] H. Yamamura, Y. Ueshima and T. Matsumiya: SCANMET II – 2nd International Conf. on Process Development in Iron and Steelmaking, 6-9 June (2004), Lulea, Sweden, 365.

[17] R. Dekkers, B. Blanpain and P. Wollants: Metall. Mater. Trans. B, 34B (2003), 161.

[18] K. Beskow and D. Sichen: Scand. J. of Metallurgy, 32 (2003), 320.

[19] K. Beskow, N. N. Viswanathan, L. Jonsson and D. Sichen: Metall. Mater. Trans. B, 32B (2001), 319.

[20] A. W. Cramb, R. Rastogi, and R. L. Maddalena: The Making, Shaping and Treating of Steel, 11th Edition, Casting Volume: The AISE Steel Foundation, Pittsburgh, PA (2003), 269.

[21] H. Goto and K. Miyazawa: ISIJ Int., 38 (1998), No. 3, 256.

[22] C. Liu, F. Huang, J. Suo, and X. Wang: Metall. Mater. Trans. B, 47B (2016), 989.

[23] J. H. Shin, Y. Chung, and J. H. Park: Metall. Mater. Trans. B, 48B (2017), 46.

[24] L. Z. Kong, Z. Y. Deng, and M. Y. Zhu: ISIJ Int., 57 (2017), 1537.

[25] S. K. Choudhary and A Ghosh: ISIJ Int., 48(2008), No. 11, 1552.

[26] K. C. Ahlborg: 5th International Clean Steel Conf, Hungary, June 1997, 15.

[27] Y. Itoh, M. Hino and S. Ban-Ya: Metall. Mater. Trans. B, 28B (1997), 953.

[28] A. Kumar, S. K. Choudhary and S. K. Ajmani: ISIJ Int., 52 (2012), 2305.

[29] G. Yang and X. Wang: ISIJ Int., 55(2015), 126.

[30] S. Dawson: Steel Times, (1992), 127.

[31] S. Dawson: Continuous Casting, ISS, Vol. 6, (1992), 53.

Influence of Ladle Addition Practice on the Castability of ...

Documents

Transcript of Influence of Ladle Addition Practice on the Castability of ...